Correspondence model-based approach for evaluating static and dynamic joint distance measurements

基于对应模型的静态和动态关节距离测量评估方法

阅读:1

Abstract

Evaluations of 3D joint space measurements between study groups have traditionally relied on surface regional divisions, which attenuate the impact of shape on joint measurements. Advancements in morphometric analyses have enabled evaluation of population-based shape variations as they relate to disease progression and deformity. Specifically, correspondence model-based shape analyses offer co-registered landmarks that address shape variability in joint structures and can be utilized for comparison of joint space measurements. This study proposes a method using correspondence models to perform group-wise statistical analyses in static or quasi-static positions during movement, offering a more comprehensive assessment of joint space variability. The primary objective was to verify and validate the measurement methods of a developed open-source toolbox. Testing was performed with surface meshes of varying edge length (0.5-, 1-, and 2-mm) and with different expected joint space distances (1- and 4-mm). Validation testing of accuracy revealed <1% error for 0.5- and 1-mm mesh edge lengths for 4 mm joint space, sensitivity testing demonstrated best results for 0.5 mm edge length, and repeatable/reliable measurements yielded low coefficient of variation and high intraclass correlation coefficient. These findings support the use of correspondence model-based approaches for robust and accurate analysis of joint measurements related to anatomical features. This method addresses limitations in traditional techniques by incorporating shape variability, providing a practical tool for assessing joint-level disease and deformity. Future work will focus on evaluating the application of this approach in diverse clinical scenarios, including highly deformed joint structures.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。